Breast cancer prediction using K Nearest Neighbors, Support Vector Machine Techniques

Abstract

Breast cancer is the trickiest and sneakiest diseases that modern science is aware of. It is one of the most important causes of death for women worldwide. We introduce the SVM (Support Vector Machine) and KNN (K Nearest Neighbors), which are the machine learning algorithms for breast disease diagnosis by training its attributes, and we present an original prediction of breast cancer. The suggested system employs 10-fold cross validation to produce accurate results. The UCI machine learning warehouse was used to obtain the Wisconsin breast melanoma diagnosis data set. The accurateness, sensitiveness, specificity, false detection rate, false exclusion rate, and Matthews' correlation coefficient are used to evaluate the presentation of the suggested system.